| dc.contributor |
Fakultas Matematika dan Ilmu Pengetahuan Alam |
|
| dc.creator |
Putri, Sharah Allysya Hajjarani |
|
| dc.creator |
Hajarisman, Nusar |
|
| dc.date |
2021-01-21 |
|
| dc.date.accessioned |
2021-03-15T03:44:56Z |
|
| dc.date.available |
2021-03-15T03:44:56Z |
|
| dc.identifier |
http://karyailmiah.unisba.ac.id/index.php/statistika/article/view/25933 |
|
| dc.identifier |
10.29313/.v7i1.25933 |
|
| dc.identifier.uri |
http://hdl.handle.net/123456789/28834 |
|
| dc.description |
Abstract. The problem in West Java province is that due to the different environmental and natural conditions in each region, the factors that cause poverty in each region must be different. Geographically Weighted Regression (GWR) is the development of a simple regression method for dealing with spatial data. This method extends the global regression model framework into a local regression model that allows local estimation of parameters. Each regression parameter is estimated at each geographic location point so that the relationship between the response variable (Y) and the predictor variable (X) varies (not the same) along the location. The weighting role of the GWR model is very important. Because this weighted value represents the location and observations of one another. One of them is kernel function. The kernel function used to estimate the parameters in the GWR model is the bi-square kernel, the bi-square weighting is an inverse function of the distance involving j (the euclidean distance between the i-th observation location and the j and b (bandwidth) observation points. The research shows the best model with a kernel weighting function bi-square seen from the smallest AIC and R-square value.Keywords : Regression, Spasial Data, Geographically Wieghted Regression (GWR), Bi-Square Weighting and Poverty.Abstrak. Permasalahan di provinsi Jawa Barat, karena kondisi lingkungan dan alam pada setiap daerah yang berbeda-beda faktor-faktor yang menyebabkan kemiskinan disetiap daerahpun pasti berbeda-beda. Geographically Weighted Regression (GWR) merupakan pengembangan dari metode regresi sederhana untuk mengatasi data spasial. Metode ini memperluas kerangka model regresi global menjadi model regresi lokal yang memungkinkan estimasi parameter secara local. Setiap parameter regresi diestimasi disetiap titik lokasi geografis sehingga hubungan antara variabel respon (Y) dan variabel prediktor (X) bervariasi (tidak sama) disepanjang lokasi. Peran pembobot pada model GWR sangatlah penting. Karena nilai pembobot ini mewakili letak dan observasi satu dengan lainnya. Salah satunya adalah fungsi kernel. Fungsi kernel yang digunakan untuk mengestimasi parameter dalam model GWR adalah kernel bi-square, pembobot bi-square merupakan fungsi invers jarak yang melibatkan dij (jarak euclidean antara titik lokasi pengamatan ke-i dengan titik pengamatan ke-j dan b (bandwidth). Hasil penelitian menunjukkan model terbaik dengan fungsi pembobot kernel bi-square dilihat dari AIC terkecil dan nilai R-squareKata Kunci : Regresi, Data Spasial, Geographically Wieghted Regression (GWR), Pembobot Bi-Square dan Kemiskinan. |
|
| dc.format |
application/pdf |
|
| dc.language |
eng |
|
| dc.publisher |
Universitas islam Bandung |
|
| dc.relation |
http://karyailmiah.unisba.ac.id/index.php/statistika/article/view/25933/pdf |
|
| dc.rights |
Copyright (c) 2021 Prosiding Statistika |
|
| dc.source |
Prosiding Statistika; Vol 7, No 1, Prosiding Statistika (Februari,2021); 208-215 |
|
| dc.source |
Prosiding Statistika; Vol 7, No 1, Prosiding Statistika (Februari,2021); 208-215 |
|
| dc.source |
2460-6456 |
|
| dc.source |
10.29313/.v7i1 |
|
| dc.subject |
Statistika |
|
| dc.subject |
Regresi, Data Spasial, Geographically Wieghted Regression (GWR), Pembobot Bi-Square dan Kemiskinan. |
|
| dc.title |
Pemodelan Data Kemiskinan di Provinsi Jawa Barat Menggunakan Metode Geographically Weighted Regression (GWR) dengan Fungsi Pembobot Kernel Bi-Square |
|
| dc.type |
info:eu-repo/semantics/article |
|
| dc.type |
info:eu-repo/semantics/publishedVersion |
|
| dc.type |
Peer-reviewed Article |
|
| dc.type |
Kuantitatif |
|